1. Field of the Invention
The present invention relates to a setting storage method in a computer system including at least one or more servers.
2. Description of the Related Art
In recent years, there has been a skyrocketing expansion in the prevalence of PC servers whose costs are lower as compared with those of mainframes and UNIX® servers. Also, the use of these PC servers is now being promoted even in mission-critical environments. This trend requires implementation of the PC servers which can exhibit high availability despite the fact that their costs are lower. What can be mentioned as the methods for enhancing the availability of a PC server are, e.g., redundant configuration of the parts, and replacement of the parts in operation. In a case where the PC server itself fails, however, the redundant configuration of the parts is meaningless. Accordingly, in a system for which the high availability is requested, redundant configuration of the PC server itself, and quick replacement and recovery of the server at the time of its failure become absolutely necessary.
Here, consideration is given to the case where the redundant configuration of servers is implemented, and where a currently-used server in operation fails. Then, when the currently-used failing server is automatically recovered back to a backup server which is prepared as the backup, a setting for the server at the point-in-time of its failure is required to be taken over to the backup server. Also, when the failing server is replaced by a new server, the setting for the server at the point-in-time of its failure is also required to be taken over to the new server after the replacement. In either case, the setting for the server at the point-in-time of its failure is required to be backed up in advance.
As the backup schemes, there have been known the scheme of making the backup manually, and the scheme of making the backup automatically. In the scheme of making the backup manually, there is the following danger: Namely, if the user forgot to make the backup after making a change in the setting, the user will find it impossible to recover the data. In view of this situation, in order to maintain the high availability, it is desirable that the server be equipped with the scheme of making the backup automatically.
The above-described setting for the server includes, e.g., a setting for the BIOS. The BIOS starts to operate with a timing with which the server boots up, and terminates when processings are over. Reading out the setting for the BIOS after these processings are over generally requires the use of a special OS or application. In many cases, however, the special OS or application like this turns out to be in an exclusive relationship with the OS or applications that should essentially be operated on the server. In view of this situation, considering from the aspect of operation's convenience, it is undesirable to carry out the backup by employing the scheme of using the special OS or application. This fact requires that, in the case of storing the setting for the BIOS, the BIOS itself back up the setting for the BIOS into another memory area using some method or other.
Several methods have been disclosed concerning the scheme of backing up the setting for the BIOS without using the special OS or application. For example, in JP-A-11-85628, the disclosure has been made concerning the scheme of periodically backing up the setting for the BIOS into a removal storage. Also, JP-A-2008-052354 and JP-A-2004-302991 (corresponding to U.S. Pat. No. 6,944,758) have disclosed the scheme of storing the setting for the backup inside the server.